Direkt zum Inhalt springen
Computer Vision Group
TUM Department of Informatics
Technical University of Munich

Technical University of Munich

Menu
Home Teaching Winter Semester 2019 Practical Course: Expert-Level Deep Learning for Computer Vision and Biomedicine (10 ECTS)

This is an old revision of the document!



Practical Course: Expert-Level Deep Learning for Computer Vision and Biomedicine (10 ECTS)

Winter Semester 2019/2020, TU München

Organizers: Vladimir Golkov, Prof. Dr. Daniel Cremers

The preliminary meeting (not obligatory) takes place on Monday, 8 July 2019 at 4pm in MI Hörsaal 3 (room 00.06.011). Slides from the preliminary meeting can be downloaded here.

Application: See below

Course Description

This course is targeted at excellent students who are on their path to being experts and making key contributions to the progress of science and technology. In this course, we will develop deep learning methods for general and specific applications in computer vision, biomedicine, and related fields. The main purpose of this course is to gain advanced practical experience and profound insights, and to learn when, why and how to apply deep learning to concrete, relevant problems. This is not limited to existing methods, but on the contrary is about the development of novel methods and solutions. Topics to choose from include the design of advanced architectures and loss functions, tasks beyond supervised learning, advanced data structures, connections to advanced non-deep/non-learning methods, etc. Experienced supervisors will guide the students and share valuable insights. Students must communicate with the supervisors.

The projects will be geared towards developing novel solutions for real open problems and contributing to scientific publications. Projects with various interesting problems and data representations will be offered.

If you want to propose an own project rather than choosing from the projects that we will offer, please discuss with us before 20 July 2019.

Prerequisites
  • Excellent programming skills
  • Eagerness to acquire and deepen knowledge about how to solve complex problems with machine learning
  • Solid skills in mathematics
  • Good literature-reading/skimming skills (Theory and practice go hand in hand!)
  • Solid knowledge of deep learning theory
  • Python and array programming in NumPy (or Matlab or similar)
  • PyTorch (or TensorFlow or similar)
  • Small gaps in the skills above are OK, because having all other skills means that gaps can be filled quickly
  • Good soft skills (or the willingness to acquire them quickly)

Choice among many different projects will be offered. Some projects will be about biomedicine. Prior knowledge of biomedicine is NOT required.

Important soft skills include communication skills, the ability to identify what is unclear, to figure out what questions need to be asked to clarify it, to formulate the questions clearly, and to ask the supervisor without hesitation. Communicating well and strategically is an important rule of the practical course. Almost all difficulties experienced by students are due to not following these rules.

Application

Step 1: Send an email to dlpractice[at]vision.in.tum.de ideally several days before you use the Matching website and at the latest when you use it and include the following:

  • Your interests, learning goals
  • Description of your skills (see prerequisites)
  • Some code you wrote in any context
  • All grade transcripts
  • Ongoing courses

Sending us an email with sufficient info about yourself is crucial for matching success.

Step 2: Apply through the TUM Matching system between 19 and 24 July 2019. Details can be found here and here.

Students who did not register or did not get matched can contact us at dlpractice[at]vision.in.tum.de

Course Structure

In the first three weeks, there will be lectures every week, focusing on theoretical and practical concepts related to deep learning. During the semester, the students will work in groups and individually on practical deep learning projects. At the end of the semester, each group will present their project with a following Q&A session. There will be no additional written or oral exam. Both the theoretical and practical part of the project will be considered in the final grading.

Course Schedule

There will be three lectures in the beginning of the semester.
Time: Tuesdays 2-4pm
Room: 03.13.010

15 October: Machine Learning; Artificial Neural Networks; Network Training; Convolutional Neural Networks; Q&A about Deep Learning
22 October: Recap; Network Architecture Design; Understanding and Visualizing; Q&A about Deep Learning
29 October: Recap; Programming; Evaluating; Q&A about Deep Learning
05 November: Q&A about Deep Learning

Note: ECTS credits are the measure of workload. So-called semester weekly hours (Semesterwochenstunden, SWS) are NOT a measure of project work time, but merely of classroom time.

Literature

Rechte Seite

Informatik IX
Chair of Computer Vision & Artificial Intelligence

Boltzmannstrasse 3
85748 Garching

info@vision.in.tum.de